Over the last 25 years, with the clinical approval of recombinant human granulocyte colony-stimulating factor (rhG-CSF) remarkable progress has been achieved in the therapy of severe congenital neutropenia (CN). Nevertheless, CN patients have a high rate of transformation to myelodysplasia (MDS) or acute myeloid leukemia (AML). This risk is especially high in the group of patients who harbor acquired G-CSFR mutations, suggesting that these mutations are involved in the leukemogenesis. Deep sequencing of the intracellular part of G-CSFR allowed us to identify a subset of CN patients with a high mutant allele frequency (MAF) of G-CSFR mutations in the granulocytic compartment. We performed CSF3R mutation analysis of myeloid colonies from colony forming unit (CFU) assay of BM and CD34+ samples of 3 ELANE -CN patients, in which the percentage of granulocytes with G-CSFR mutation varied between 22.6% and 81%. Average number of sequenced colonies per CFU sample was 18. Intriguingly, we found that only cell clones with WT G-CSFR were grown in CFU assay. We hypothesized that lack of G-CSFR mutant clones in CFU assay is due to abnormal signaling downstream of truncated G-CSFR. To compare myeloid differentiation of G-CSFR mutant and WT G-CSFR hematopoietic cells, we generated ELANE -CN patient-derived induced pluripotent stem cells (CN-iPSCs) and performed CRISPR/Cas9 genome editing of CSF3R . Using 2 different sgRNAs specific for the intracellular region of G-CSFR, we introduced nucleotide indels at amino acid positions 740 and 735 (NP_000751.1) in CN-iPSCs. Subsequently, homozygous and heterozygous CN-iPSCs clones with truncated distal part of cytoplasmic domain of G-CSF receptor and subsequent loss of 3 out of 4 conserved tyrosine residues were generated.

We compared myeloid differentiation of healthy donor iPSCs (HD-iPSCs), CN-iPSCs and CN-iPSCs with heterozygous and homozygous truncated G-CSF receptor (G-CSFR*HET CN-iPSCs and G-CSFR*HOM CN-iPSCs, respectively). Absolute cell count during myeloid differentiation (18-32 day) of HD-iPSCs, G-CSFR*HET CN-iPSCs and G-CSFR*HOM CN-iPSCs didn't differ significantly (p>0.05). In contrast, absolute cell count was found to be elevated from day 25 to day 32 of differentiation when G-CSFR*HET CN-iPSCs and G-CSFR*HOM CN-iPSCs hematopoietic progenitors were compared with CN-iPSCs-derived hematopoietic cells (p<0.05). FACS analysis of CN-iPSCs derived hematopoietic progenitors with and without truncated G-CSF receptor using a panel of hematopoietic markers (KDR, CD34, CD235a, CD41a, CD45 and CD33) showed no difference in percentage of early progenitors (CD34+/KDR+), E/MK precursors (CD41a+CD235a+/ CD45-), CD34+/CD45+ and CD34+/CD43+ cells. Intriguingly, we observed increase in percentage of CD34+/CD43+ (p<0.01), CD33+/CD45+ (p<0.05) and CD41a+CD235a+/ CD45- (p<0.05) cells at day 14 of myeloid differentiation of G-CSFR*HET CN-iPSCs and G-CSFR*HOM CN-iPSCs versus HD-iPSc. As expected, relative number CD14+ and CD16+ cells were not changed upon myeloid differentiation of hematopoietic progenitors with mutant G-CSFR but relative levels of CD15+ and CD15+/CD16+ were negatively affected by homozygous and heterozygous G-CSFR mutations. Next, CFU assay of hematopoietic progenitors derived from G-CSFR*HET CN-iPSCs and G-CSFR*HOM CN-iPSCs produced markedly less colonies than HD- and CN-iPSCs derived hematopoietic cells. Cell morphology analysis also confirmed the reduced number of mature granulocytes seen after myeloid differentiation of G-CSFR*HET CN-iPSCs and G-CSFR*HOM CN-iPSCs. Finally, reprogramming of BM cells from CN patients into iPSCs followed by mutation analysis of single cell derived iPSCs clones allowed us to identify iPSCs with naturally acquired G-CSFR mutations. Preliminary results revealed reduced myeloid differentiation of these cells, which is in line with data generated from G-CSFR*HET CN-iPSCs and G-CSFR*HOM CN-iPSCs.

Using a combination of patient CN-iPSCs and CRISPR/Cas9 gene editing we were able to recapitulate main features of G-CSFR mutant versus G-CSFR WT CN cells - defective neutrophil differentiation and increased proliferation of human hematopoietic progenitors. Since virtually all studies of truncated G-CSFR receptor signaling have been performed in murine cytokine-responsive myeloid cell lines, our approach is more suitable to explore human aberrant G-CSF receptor signaling in depth.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.